Abstract: Objective To investigate the risk factors of cataract in the elderly patients with uveitis, and to establish a nomogram predictive model. Methods A total of 223 patients with uveitis admitted to Union Hospital, Tongji Medical College, Huazhong University of Science and Technology from March 2021 to March 2024 were retrospectively selected. Relevant clinical data were collected. All patients were divided into cataract group (116 cases) and non-cataract group (107 cases) based on the presence of cataract complications. Multivariate logistic regression analysis was conducted to analyze the related factors affecting cataract complications in the elderly patients with uveitis, and the prediction model was constructed and verified by nomogram. Results Higher disease severity score, the frequency of uveitis ≥2 times per year, cystoid macular edema (CME), diabetes, long-term use of glucocorticoids, older age and family history of cataract were identified as the risk factors for cataract in the elderly patients with uveitis (P < 0.05). The regression-based nomogram predictive model showed good calibration (mean absolute error=0.02) and discrimination (C-index=0.851). Its predictive performance was also excellent, with an area under the receiver operating characteristic curve of 0.885. Conclusions The incidence rate of cataract is high in the elderly patients with uveitis. Risk factors include advanced age, family history of cataract, severity of uveitis, diabetes and long-term use of hormones, and concomitant CME. The nomogram model constructed based on these factors shows high predictive value for cataract in the elderly patients with uveitis.

Key words: uveitis, risk factors, predictive model, cataract